Immunocompromised patients are susceptible to a variety of neoplastic, protozoal, viral, bacterial and fungal diseases; of these, bacterial, viral and fungal infections result in the greatest mortality (Bartlett, M. and J. Smith, Clin. Microbiol. Rev. 4:137 (1991); Bodey, G. et al., Dur. J. Clin. Microbiol. Infect. Dis. 11:99 (1992); Sternberg, S., Science 266:1632 (1994); Cox, G. and J. Perfect, Curr. Opin. Infect. Dis. 6:422 (1993); Deepe, G. and W. Bullock, Dur. J. Clin. Microbiol. Infect. Dis. 9:377 (1990); Fox, J. L., ASM News 59:515 (1993); Kujath, P., Mycoses 35:225 (1992); Pfaller, M. and R. Wenzel, Dur. J. Clin. Microbiol. Infect. Dis. 11:287 (1992); and Samonis, G. and D. Bafaloukos, In vivo 6:183 (1992)). Systemic fungal infections are now an important medical problem in the care of immunocompromised patients. Fungal infections in immunocompromised patients are mainly the result of opportunistic infections by normally harmless, asymptomatic commensals, which can be pathogenic under certain conditions (Odd, J., Antimicrob. Chemother. 31:463 (1993); Rhodes, J. et al., J. Med. Vet. Myc. 20:113 (1992); Saral, R. Rev. Infect. Dis. 13:487 (1991)). Species of Cryptococcus, Candida, Coccidioides, Histoplasma, Blastomyces, Sporothrix and Aspergillus, as well as other opportunistic fungi, are important causative agents; of these, Candida species, especially C. albicans, are the most common. Candidemia accounts for 8-10% of all hospital-acquired bloodstream infections and Candida species are the fourth most common cause of nosocomial septicemias. Mortality rates associated with systemic Candida infections are estimated to be as high as 50% of infected patients. Infections caused by other types of fungi (e.g., Aspergillus, Cryptococcus) are also common in immunocompromised patients and carry significant treatment costs and mortality levels (Meunier, F., Amer. J. Med. 99 (Suppl. 6A):60S-67S (1995)).
A variety of approaches have been used to diagnose fungal infections in humans; each method has distinct limitations. Blood cultures often fail to detect existing infections and may take two to five days to detect fungal growth, a period of time which an infected patient often cannot survive. Assays for fungal-specific metabolites are also used. For example, an assay for D-arabinitol can be performed by monitoring of oxidation of D-arabinitol by D-arabinitol dehydrogenase (Soyama, K. and E. Ono, Clin. Chim. Acta 149:149 (1985); Soyama, K. and E. Ono, Clin. Chim. Acta. 168:259 (1987)). However, D-arabinitol dehydrogenase also reacts with D-mannitol naturally existing in serum samples, resulting in an artificially high oxidation rate. Furthermore, both D-mannitol and D-arabinitol are elevated during renal failure (Reiss, E. and C. Morrison, J. Clin. Microbiol. Rev. 6:311-322 (1993)).
Immunoassays to detect antibody binding to a fungal marker antigen have also been developed: for example, an assay for enolase secreted by C. albicans is described (Matthews, R. C. et al., Lancet i:1415 (1984), Matthews, R. C. et al., J. Clin. Microbiol. 26:459 (1988), and Walsh, et al., New. Engl. J. Med. 324:1026 (1991)). However, the capture antibody is believed to react with enolases from sources other than C. albicans (Reiss, E. and C. Morrison, J. Clin. Microbiol. Rev. 6:311-322 (1993), and the sensitivity is limited (Mitsutake, K. et al., J. Clin. Microbiol. 34:1918-1921 (1996)). Another assay uses detection of aspartyl proteinase secreted from C. albicans (Staib, R., Sabouraudia 4:187 (1965); Reiss, E. and C. J. Morrison, Clin. Microbiol. Rev. 6:311 (1993)). An assay to detect secreted .beta.(1-3)-glucan, based on activation of factor G of the Limulus coagulation enzyme cascade, is also described (U.S. Pat. No. 5,266,461); however, this assay is sensitive to a number of interfering compounds, including compounds of fungal origin (Matsumoto, T. M. et al., Urol. Res. 21:21-117-120 (1993)).
Additional assays include immunological assays to detect antigens, such as cell-wall mannan, circulating through the host during fungal infection, by radioimmunassay (RIA) (Weiner, M. H. and M. Coats-Stephen, J. Infect. Dis. 140:989 (1979)), inhibition enzyme immunoassay (meckstroth, K. L. et al., J. Infect. Dis. 144:24 (1981); Segal, E. et al., J. Clin. Microbiol. 10:116-118 (1979)), and double antibody sandwich enzyme immunoassay (Lew, M. A. et al., J. Infect. Dis. 145:45-56 (1982)). Reliability of these tests may be hampered by the presence of anti-mannan antibodies and cell-wall binding proteins. Furthermore, the concentration of the antigen is low, even with severe infection, so detection can be difficult.
Amplification of sample DNA by polymerase chain reaction (PCR) has been used with fungal-specific nucleotide probes to detect the presence of fungal DNA (U.S. Pat. Nos. 5,489,513; 5,426,027; and 5,324,632). These methods are complicated, expensive, time-consuming and often are sensitive to interfering materials found in the host sample (Reiss, E. and C. Morrison, J. Clin. Microbiol. Rev. 6:311-323 (1993)).
Thus, each of the variety of methods for diagnosing fungal infection has deficiencies, including unreliability, interfering agents, or a need for sophisticated equipment or trained personnel. Because of the increasing incidence of fungal infections, a need remains for accurate, simple, quantitative and expedient methods for diagnosis of fungal infections.